Re: [SQLObject] PostgreSQL: costly count

[Frank B. <fb...@fo...>]

Hallo,
J-P Lee hat gesagt: // J-P Lee wrote:

> This is a common problem with most db's.  select count(id) is usually 
> the prefered way to do it. (Although I haven't tested the speed diff).

I did,  it's largely the same as count(*) on Postgres, and also uses a
sequential scan through the whole DB. I'm just reading through this
looong thread on pgsql-performance:
http://archives.postgresql.org/pgsql-performance/2003-10/msg00058.php

ciao
-- 
 Frank Barknecht                               _ ______footils.org__